WO2022172882A1 - 推測装置、推測システム、推測プログラム及び推測方法 - Google Patents

推測装置、推測システム、推測プログラム及び推測方法 Download PDF

Info

Publication number
WO2022172882A1
WO2022172882A1 PCT/JP2022/004611 JP2022004611W WO2022172882A1 WO 2022172882 A1 WO2022172882 A1 WO 2022172882A1 JP 2022004611 W JP2022004611 W JP 2022004611W WO 2022172882 A1 WO2022172882 A1 WO 2022172882A1
Authority
WO
WIPO (PCT)
Prior art keywords
water system
parameter
parameters
relationship model
water
Prior art date
Application number
PCT/JP2022/004611
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
康広 豊岡
昌樹 田頭
仁樹 桂
要 原田
勝彦 日▲高▼
Original Assignee
栗田工業株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 栗田工業株式会社 filed Critical 栗田工業株式会社
Priority to KR1020237027836A priority Critical patent/KR20230133886A/ko
Priority to EP22752706.6A priority patent/EP4286583A1/en
Priority to CN202280014302.7A priority patent/CN116829785A/zh
Publication of WO2022172882A1 publication Critical patent/WO2022172882A1/ja

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/34—Paper
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F1/00—Treatment of water, waste water, or sewage
    • C02F1/008—Control or steering systems not provided for elsewhere in subclass C02F
    • D—TEXTILES; PAPER
    • D21—PAPER-MAKING; PRODUCTION OF CELLULOSE
    • D21F—PAPER-MAKING MACHINES; METHODS OF PRODUCING PAPER THEREON
    • D21F1/00—Wet end of machines for making continuous webs of paper
    • D21F1/66—Pulp catching, de-watering, or recovering; Re-use of pulp-water
    • D—TEXTILES; PAPER
    • D21—PAPER-MAKING; PRODUCTION OF CELLULOSE
    • D21F—PAPER-MAKING MACHINES; METHODS OF PRODUCING PAPER THEREON
    • D21F7/00—Other details of machines for making continuous webs of paper
    • D21F7/02—Mechanical driving arrangements
    • D—TEXTILES; PAPER
    • D21—PAPER-MAKING; PRODUCTION OF CELLULOSE
    • D21G—CALENDERS; ACCESSORIES FOR PAPER-MAKING MACHINES
    • D21G9/00—Other accessories for paper-making machines
    • D21G9/0009—Paper-making control systems
    • D—TEXTILES; PAPER
    • D21—PAPER-MAKING; PRODUCTION OF CELLULOSE
    • D21J—FIBREBOARD; MANUFACTURE OF ARTICLES FROM CELLULOSIC FIBROUS SUSPENSIONS OR FROM PAPIER-MACHE
    • D21J7/00—Manufacture of hollow articles from fibre suspensions or papier-mâché by deposition of fibres in or on a wire-net mould
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/18—Water
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
    • G01N33/18—Water
    • G01N33/1806—Biological oxygen demand [BOD] or chemical oxygen demand [COD]
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2103/00—Nature of the water, waste water, sewage or sludge to be treated
    • C02F2103/26—Nature of the water, waste water, sewage or sludge to be treated from the processing of plants or parts thereof
    • C02F2103/28—Nature of the water, waste water, sewage or sludge to be treated from the processing of plants or parts thereof from the paper or cellulose industry
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/001—Upstream control, i.e. monitoring for predictive control
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/003—Downstream control, i.e. outlet monitoring, e.g. to check the treating agents, such as halogens or ozone, leaving the process
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/005—Processes using a programmable logic controller [PLC]
    • C02F2209/006—Processes using a programmable logic controller [PLC] comprising a software program or a logic diagram
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/02—Temperature
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/04—Oxidation reduction potential [ORP]
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/05—Conductivity or salinity
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/05—Conductivity or salinity
    • C02F2209/055—Hardness
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/06—Controlling or monitoring parameters in water treatment pH
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/08—Chemical Oxygen Demand [COD]; Biological Oxygen Demand [BOD]
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/09—Viscosity
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/10—Solids, e.g. total solids [TS], total suspended solids [TSS] or volatile solids [VS]
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/10—Solids, e.g. total solids [TS], total suspended solids [TSS] or volatile solids [VS]
    • C02F2209/105—Particle number, particle size or particle characterisation
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/11—Turbidity
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/23—O3
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/26—H2S
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/29—Chlorine compounds
    • C—CHEMISTRY; METALLURGY
    • C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
    • C02F2209/00—Controlling or monitoring parameters in water treatment
    • C02F2209/36—Biological material, e.g. enzymes or ATP

Definitions

  • the present invention relates to a guessing device, guessing system, guessing program, and guessing method.
  • Patent Document 1 proposes a method of measuring two or more water quality parameters related to water quality in a water system for manufacturing paper and performing water treatment of the water system based on the measured values. And according to the method shown in this patent document 1, high-quality paper can be manufactured.
  • Patent Document 1 does not disclose the occurrence of troubles in the water system used to manufacture paper and the quantitative estimation of the quality of paper products. On the other hand, it is necessary to quantitatively estimate the occurrence of troubles and the quality of paper products in the water system in order to control the operating conditions and the amount of chemicals added to the system more precisely.
  • the present invention provides an estimation device, an estimation system, an estimation program, and an estimation method that can quantitatively estimate the occurrence of troubles in water systems and the quality of products manufactured through water systems. .
  • an inferring device for inferring probable future outcomes in or derived from a water system.
  • This estimating device includes a parameter information acquiring section, a relationship model information acquiring section, and an estimating section.
  • the parameter information acquisition unit acquires water quality parameters related to the water quality of the water system, control parameters related to the control conditions of equipment related to the water system or raw materials added to the water system, and parameters that have different meanings from the expected results and are related to the water system and the water system Parameter information that includes two or more parameters that are any one of the result parameters related to the results generated in the equipment or raw materials added to the water system, or derived from the water system, the equipment related to the water system, or the raw materials added to the water system get.
  • the relationship model information acquisition unit acquires relationship model information indicating a relationship between an expected result or an index related to the expected result and two or more parameters created in advance.
  • the estimation unit estimates an expected result or an indicator related to the expected result based on the parameter information and the relationship model information.
  • the relationship model is a regression analysis, a time series analysis, a decision tree, and a prior confirmation result corresponding to the expected result or an index related to the prior confirmation result and two or more of the parameters.
  • a guesser that is a model determined by neural network, Bayesian, clustering or ensemble learning.
  • the water system is a water system in the process of manufacturing paper products.
  • the water quality parameters are pH, electrical conductivity, redox potential, zeta potential, turbidity, temperature, foam height, biochemical oxygen demand (BOD), chemical oxygen demand of the water system.
  • control parameters are the operating speed (papermaking speed) of the paper machine, the rotation speed of the filter cloth of the raw material dehydrator, the rotation speed of the filter cloth of the washer, the amount of chemical added to the water system, and the raw material to be added to the water system.
  • Amount of chemical added to water system, amount of chemical added to equipment related to the water system, amount of steam for heating, steam temperature for heating, steam pressure for heating, flow rate from seed box, nip pressure of press part, felt of press part One type selected from the group consisting of vacuum pressure, blending ratio of raw materials for paper manufacturing, waste paper blending amount of raw materials for manufacturing paper, opening of screen for raw materials for manufacturing paper, gap distance between rotor and stator of beater, freeness and degree of beating.
  • a guessing device that is more than that.
  • the result parameters are the unit weight of the paper product (weight per square meter), the yield rate, the concentration of white water, the moisture content of the paper product, the amount of steam in the equipment for manufacturing the paper product, and the paper product.
  • An estimation system for estimating expected results that may occur in the water system or derived from the water system comprising a parameter information acquisition unit, a relationship model information acquisition unit, and an estimation unit, wherein the parameter information acquisition unit is a water quality parameter related to the water quality of the water system, a control parameter related to the water system, equipment related to the water system, or a control condition of raw materials added to the water system, and a parameter having a meaning different from the expected result, wherein the water system, the A parameter that is any one of a result parameter related to a result that occurs in a facility associated with a water system or a raw material added to said water system, or derived from said water system, equipment associated with said water system, or raw material added to said water system.
  • the relationship model information acquisition unit creates a relationship indicating a relationship between the expected result or an index related to the expected result created in advance and the two or more parameters
  • An inference system that acquires model information, and in which the estimation unit estimates the expected result or an index related to the expected result based on the parameter information and the relationship model information.
  • An estimation program for estimating expected results that may occur in the water system or derived from the water system wherein the computer functions as a parameter information acquisition unit, a relationship model information acquisition unit, and an estimation unit, and the parameter information acquisition
  • the part is a water quality parameter related to the water quality of the water system, a control parameter related to the control condition of the water system, equipment related to the water system, or a raw material to be added to the water system, and a parameter having a meaning different from the expected result, the water system,
  • a parameter that is any one of result parameters relating to a result generated in equipment associated with said water system or raw materials added to said water system or derived from said water system, equipment associated with said water system or raw materials added to said water system and the relationship model information acquisition unit creates a relationship indicating the relationship between the expected result or an index related to the expected result and the two or more parameters
  • a guessing program that acquires sex model information, and the guessing unit guesses the expected result or an index related to the expected result based on the parameter information
  • An estimation method for estimating expected results that may occur in a water system or derived from the water system comprising a parameter information acquisition step, a relationship model information acquisition step, and an estimation step, wherein the parameter information acquisition step.
  • a water quality parameter related to the water quality of the water system a control parameter related to the control conditions of the water system, equipment related to the water system, or raw materials to be added to the water system, and a parameter having a meaning different from the expected result
  • the water system the A parameter that is any one of a result parameter relating to a result that occurs in equipment related to a water system or raw materials added to said water system or derived from said water system, equipment related to said water system or raw materials added to said water system.
  • Acquiring parameter information including two or more, and in the relationship model information acquiring step, a relationship indicating a relationship between the expected result or an index related to the expected result created in advance and two or more of the parameters
  • An estimation method for acquiring model information and, in the estimation step, estimating the expected result or an indicator related to the expected result based on the parameter information and the relationship model information is not limited to this.
  • FIG. 1 is a schematic diagram of an inference system according to this embodiment;
  • FIG. It is a schematic diagram which shows the functional structure of the inference apparatus which concerns on this embodiment.
  • It is a schematic diagram showing the hardware constitutions of the guessing device concerning this embodiment.
  • 10 is a plot of the number of trouble occurrences A for a total of 30 sets of data sets versus an index a related to the likely outcome;
  • FIG. 10 is a graph showing the degree of influence of each parameter on index a related to expected results;
  • 1 is a schematic diagram of equipment for manufacturing paper in Example 1.
  • FIG. 4 is a plot of the number of defects versus defect index ⁇ for a total of 572 data sets for Example 1; 7 is a graph showing the degree of influence of each parameter on the defect index; 4 is a plot of strength agent usage intensity vs. strength index value for a total of 60 data sets for Example 2; 4 is a graph showing the degree of influence of each parameter on the paper strength index; 4 is a graph showing the occurrence of paper breakage and changes over time in the paper breakage index. 4 is a graph showing the occurrence of paper breakage and changes over time in the paper breakage index.
  • FIG. 4 is a schematic diagram of equipment for manufacturing paper in Example 4.
  • FIG. 11 is a plot of the number of defects versus defect index ⁇ values for a total of 647 sets of data sets for relationship modeling in Example 4;
  • FIG. 11 is a plot of the number of defects versus defect index ⁇ values for a total of 255 sets of data sets for accuracy verification in Example 4;
  • FIG. 10 is a plot showing changes over time in defect indexes calculated from a total of 255 sets of data sets for accuracy verification in Example 4.
  • FIG. 4 is a graph showing the degree of influence of each parameter on defect index ⁇ .
  • FIG. 11 is a plot of the number of defects versus the defect index ⁇ value for a total of 631 sets of data sets for relationship modeling in Example 5;
  • FIG. 10 is a plot of the number of defects versus defect index ⁇ values for a total of 255 sets of data sets for accuracy verification in Example 5; 10 is a plot showing changes over time in defect indexes calculated from a total of 271 sets of data sets for accuracy verification in Example 5.
  • FIG. 4 is a graph showing the degree of influence of each parameter on defect index ⁇ .
  • FIG. 11 is a plot of the number of defects versus the defect index ⁇ value for a total of 1216 sets of data sets for relationship modeling in Example 6;
  • FIG. 10 is a plot of the number of defects versus the defect index ⁇ value for a total of 490 sets of data sets for accuracy verification in Example 6; 4 is a graph showing the degree of influence of each parameter on defect index ⁇ .
  • FIG. 11 is a plot of the number of defects versus the defect index ⁇ value for a total of 1216 sets of data sets for relationship modeling in Example 6;
  • FIG. 10 is a plot of the number of defects versus the defect index ⁇ value for
  • FIG. 11 is a plot of defect count versus defect index ⁇ for a total of 1503 sets of data sets for relationship modeling in Example 7;
  • FIG. 11 is a plot of the number of defects versus defect index ⁇ for a total of 537 sets of data sets for accuracy verification in Example 7;
  • 4 is a graph showing the degree of influence of each parameter on defect index ⁇ .
  • the program for realizing the software of this embodiment may be provided as a computer-readable non-transitory recording medium (Non-Transitory Computer-Readable Medium), or may be provided downloadably from an external server. Alternatively, it may be provided so that the program is activated by an external computer and the function is realized by the client terminal (so-called cloud computing).
  • the term “unit” may include, for example, a combination of hardware resources implemented by circuits in a broad sense and software information processing that can be specifically realized by these hardware resources.
  • various information is handled in the present embodiment, and these information are, for example, physical values of signal values representing voltage and current, and signal values as binary bit aggregates composed of 0 or 1. It is represented by high and low, or quantum superposition (so-called quantum bit), and communication and operation can be performed on a circuit in a broad sense.
  • a circuit in a broad sense is a circuit realized by at least appropriately combining circuits, circuits, processors, memories, and the like.
  • Application Specific Integrated Circuit ASIC
  • Programmable Logic Device for example, Simple Programmable Logic Device (SPLD), Complex Programmable Logic Device (CPLD), and field It includes a programmable gate array (Field Programmable Gate Array: FPGA)).
  • the estimating system is an estimating system for estimating possible future outcomes in or derived from a water system.
  • this estimation system includes a parameter information acquisition section, a relationship model information acquisition section, and an estimation section.
  • the parameter information acquisition unit obtains a water quality parameter related to the water quality of the water system, a control parameter related to the control condition of the water system, equipment related to the water system, or the raw material to be added to the water system, and a parameter having a meaning different from the expected result and the water system
  • Two or more parameters that are any one of the following: , result parameters related to results generated in equipment related to water systems or raw materials added to water systems, or derived from water systems, equipment related to water systems, or raw materials added to water systems It acquires parameter information including
  • the relationship model information acquisition unit acquires relationship model information indicating a relationship between an expected result or an index related to the expected result created in advance and two or more parameters.
  • the estimation unit estimates an expected result or an index related to the expected result based on the parameter information and the relationship model information.
  • the inference system may include one or both of the relationship model creation unit and the output unit. Note that FIG. 1, which will be described below, mainly describes an inference system that includes all of these.
  • FIG. 1 is a schematic diagram of an inference system according to this embodiment.
  • This guessing system 1 comprises a guessing device 2 and an output device 3 .
  • FIG. 2 is a schematic diagram showing the functional configuration of the guessing device according to this embodiment.
  • the estimating device 2 according to the present embodiment is an estimating device for estimating expected results that may occur in the future in the water system W or derived from the water system W, and includes a parameter information acquisition unit 21, A relationship model information acquisition unit 22 and an estimation unit 23 are provided.
  • the estimation device 2 according to this embodiment further includes a second estimation unit 24 , a relationship model creation unit 25 , a second relationship model creation unit 26 , and a second relationship model information acquisition unit 27 . Although each of these units is described here as being included inside one device, each unit may be included as a separate device.
  • the output device 3 is an example of an output unit
  • the parameter information measurement device 4 is an example of a parameter information measurement unit.
  • expected results that may occur in the water system or derived from the water system are the results that will occur in the water system when the water system is operated.
  • BOD biochemical oxygen demand
  • COD increase in
  • SS suspended solids
  • turbidity increase increase in chromaticity increase
  • transparency decrease dehydrated sludge amount increase
  • heat exchanger efficiency decrease chiller efficiency decrease
  • chiller efficiency decrease Reduced efficiency of cooling towers, increased frequency of backwashing filter media and activated carbon, increased frequency of replacement of filter media and activated carbon, increased frequency of cleaning membranes (MF membrane, UF membrane, RO membrane, etc.), Increased replacement frequency, increased regeneration frequency of ion exchange resin, increased replacement frequency of ion exchange resin, corrosion of equipment and piping (due
  • expected results that may occur in the future derived from water systems are the results that occur in relation to water systems other than those water systems when such water systems are operated and operated. performance deterioration, product yield reduction, unwanted by-products increase, and changes in product odor.
  • the "index related to the expected result” may be any index that has a certain correlation with the expected result (for example, a function of two or more parameters), and is not a generally known index but a speculation It may be created independently by the system user (operator, etc.).
  • the "expected results that may occur in the future in the water system" of the paper manufacturing process include, for example, contamination of the paper machine, contamination of the papermaking approach system, contamination of the white water recovery system, pump air entrapment, screen blockage, reduction in papermaking speed, wire part Poor drainage, poor dehydration in the press part, poor drying in the dryer part, bad odor in the water system, poor peeling in the dryer process, dirt on the block system, dirt on the raw material system, and the like.
  • the "expected results that may occur in the future derived from the water system" of the paper manufacturing process are, for example, those related to paper products produced from such water system (number of defects, paper strength, joint ratio, sizing degree, air permeability, smoothness degree, ash content, color tone, whiteness, formation, odor, causticization rate, firing rate, kappa number, freeness, moisture content, etc.), and events that can occur outside of water systems (paper breakage in the press part to dryer part, etc.) is mentioned.
  • the parameter information acquisition unit 21 obtains a water quality parameter related to the water quality of the water system W, a control parameter related to the control condition of the water system W, equipment related to the water system W, or a raw material to be added to the water system W, and a parameter having a different meaning from the expected result, Any one of the result parameters related to the result generated in the water system W, the equipment related to the water system W, or the raw material added to the water system W, or derived from the water system W, the equipment related to the water system W, or the raw material added to the water system W This is to acquire parameter information including two or more parameters.
  • the water system here is not limited to those existing in one tank or channel or those in which a continuous flow exists, but those having a plurality of tanks or channels, specifically, A water system in which branching or confluence of a plurality of flow paths exists, or in which water is transferred from tank to tank in units of batches, or in which treatment is performed in the middle, is also considered as one water system.
  • the water system related to the water quality parameters, control parameters or result parameters is divided by tanks, etc., the water quality parameters, control parameters or result parameters for a part of the water system may be used, and the water quality parameters for the entire water system may be used. , control parameters or outcome parameters may be used.
  • the water quality parameter is not particularly limited as long as it relates to the water quality of the water system W.
  • the control parameter is not particularly limited as long as it relates to the control condition of the water system W, equipment related to the water system W, or raw material added to the water system W.
  • a parameter that has a different meaning from the expected result and is derived from the water system W, the equipment related to the water system W, or the raw material added to the water system W, or from the water system W, the equipment related to the water system W, or the raw material added to the water system W It is not particularly limited as long as it relates to the results produced by Note that “has a different meaning from the expected result” means that if the evaluation index (e.g., physical quantity) is different (e.g., one is length and the other is mass), the evaluation index is the same but the evaluation results are different. Including cases where the target is different (e.g. the mass of paper and the mass of additives) and cases where the measurement points are different (e.g.
  • Water quality parameters include, for example, water system pH, electrical conductivity, redox potential, zeta potential, turbidity, temperature, foam height, biochemical oxygen demand (BOD), chemical oxygen demand (COD), absorbance (e.g., UV absorbance), color (e.g., RGB value), particle size distribution, degree of aggregation, amount of foreign matter, foamed area on water surface, soiled area in water, amount of air bubbles, amount of glucose, amount of organic acid, amount of starch , the amount of calcium, the amount of total chlorine, the amount of free chlorine, the amount of dissolved oxygen, the amount of cation demand, the amount of hydrogen sulfide, the amount of hydrogen peroxide, and the respiration rate of microorganisms in the system. It is preferable to use the above.
  • Control parameters include, for example, the operating speed (papermaking speed) of the paper machine, the rotation speed of the filter cloth of the raw material dehydrator, the rotation speed of the filter cloth of the washer, the amount of chemical added to the water system, the amount of chemical added to the raw material added to the water system, the water system Amount of chemical added to the equipment related to , Steam volume for heating, Steam temperature for heating, Steam pressure for heating, Flow rate from seed box, Press part nip pressure, Press part felt vacuum pressure, Mixing of papermaking raw materials It is preferable to use one or more selected from the group consisting of ratio, waste paper blending amount of papermaking raw material, screen opening of papermaking raw material, gap distance between rotor and stator of beating machine, freeness and beating degree.
  • the water system W is a water system in the process of manufacturing paper products
  • equipment related to the water system include equipment such as paper machine wires and felts that directly add chemicals.
  • Result parameters include, for example, the unit weight of paper products (weight per square meter), yield rate, white water concentration, moisture content of paper products, amount of steam in the facility that manufactures paper products, and amount of steam in the facility that manufactures paper products. , steam temperature in equipment for manufacturing paper products, steam pressure in equipment for manufacturing paper products, thickness of paper products, ash concentration in paper products, types of defects in paper products, number of defects in paper products, It is preferable to use one or more selected from the group consisting of time of paper breakage in the process, freeness, beating degree and aeration amount.
  • the amount of steam in equipment for manufacturing paper products includes, for example, the amount of steam in paper machine dryers, the amount of steam in kraft pulp black liquor evaporators, the amount of steam in black liquor heaters in kraft pulp digesters, the amount of steam in pulp raw materials and white water Steam volume blowing for temperature can be used.
  • two or more parameters acquired by the parameter information acquisition unit should not be substantially the same.
  • the two parameters are:
  • the use of process steam generated from black liquor as a result parameter and the amount of steam used to warm (concentrate) concentrated black liquor as a control parameter shall be excluded. This is because in such a case, the process steam generated from the black liquor as the result parameter and the amount of steam used for warming (concentrating) the concentrated black liquor as the control parameter are substantially the same.
  • the two parameters are the process steam generated from black liquor as a result parameter and , and the amount of steam used for warming (concentrating) the concentrated black liquor as control parameters. This is because in such cases, the process steam generated from the black liquor as the result parameter and the amount of steam used to warm (concentrate) the concentrated black liquor as the control parameter are not substantially the same.
  • the two parameters are the process steam generated from the black liquor as a result parameter , and the amount of steam used for heating (concentrating) the concentrated black liquor as a control parameter, in addition to other parameters such as the pH of the water system as a water quality parameter. may be substantially the same.
  • freeness and freeness are the same parameters, but can be included in both the control parameters and the result parameters.
  • the water quality parameter, control parameter, and result parameter are concepts that each include multiple parameters.
  • parameter information including two or more parameters two or more parameters included in the parameter information can be independently selected from the water quality parameters, the control parameters, and the result parameters.
  • a combination of two or three of the water quality parameters, the control parameters and the result parameters e.g. pH and the press pressure and thickness of the paper product.
  • exactly the same parameter for example, the pH of water at location A and the pH of water at location A
  • should not be selected (however, for example, the pH of water at location A and the water at location B, which are different measurement locations) ).
  • the relationship model information acquisition unit 22 acquires relationship model information indicating the relationship between an expected result or an index related to the expected result, which has been created in advance, and two or more parameters.
  • a relationship model is created in advance and shows the relationship between expected results or indicators related to expected results and two or more parameters.
  • "preliminary" refers to the expected result or before estimating the index related to the expected result, and whether the expected result or the expected result is Any prior to estimating indices related to .
  • the relationship model is not particularly limited, but for example, an expected result or an index related to the expected result, a function indicating the relationship between two or more parameters, a lookup table or an expected result or an index related to the expected result, Examples include a trained model of the relationship between two or more parameters.
  • Two or more parameters included in the parameter information acquired by the parameter information acquisition unit 21 and parameters included in the parameter information used in the relationship model shall have two or more in common.
  • the water quality parameter, control parameter, and result parameter are concepts that each include a plurality of parameters. "Two or more of the parameters are common" means that two or more of the water quality parameters, control parameters, and result parameters (e.g., two or more only water quality parameters; water pH and temperature) are common.
  • a combination of water quality parameters, control parameters, and result parameters may be common, or water quality parameters, All control parameters and result parameters (eg, one water quality parameter, one control parameter and one result parameter) may be common.
  • the estimation unit 23 estimates an expected result or an index related to the expected result based on the parameter information and the relationship model information.
  • the estimating unit 23 inputs the current parameter information of the water system W into the relationship model created in advance, substitutes or compares it with the relationship model, and obtains an expected result or an index related to the expected result. to estimate (calculate).
  • the second estimation unit 24 estimates an expected result from the index when the estimation unit 23 estimates an index related to the expected result instead of the expected result itself.
  • the estimation unit 23 is called a "first estimation unit" for convenience.
  • the first estimation unit 23 estimates an index related to the expected result (hereinafter also referred to as "related indicator"), it is necessary to estimate the expected result from the related indicator. Specifically, this related index is input to a second relationship model prepared in advance, and an expected result is estimated. In addition, when using a 2nd relationship model, the relationship model used by the 1st estimation part 23 is called a “1st relationship model" for convenience.
  • a threshold can be set for the related index, and it can be inferred that trouble will occur if the related index is greater (or smaller) than the threshold.
  • the related indicator When estimating the possibility of trouble occurring, for example, by setting multiple thresholds for the related indicators, for example, dividing them into three stages, when the related indicator is in the first stage, the occurrence of trouble will certainly occur, and the related indicator will When it is in the second stage, it can be assumed that trouble may occur, and when the related index is in the third stage, it can be inferred that trouble will not occur. Also, the relationship between the related index and the probability of occurrence of trouble can be converted into a function or a learned model from the statistical data of actual operation, and the probability of occurrence of trouble can be calculated.
  • the relationship model creating unit 25 creates a relationship model. This relationship model may be acquired by the relationship model information acquisition unit 22 and used by the estimation unit 23 to estimate the expected result or an index related to the expected result.
  • a relationship model is created, for example, as follows. Prior to inferring a prospective outcome or related metric, a pre-measured outcome corresponding to the prospective outcome or a pre-measured metric associated with the pre-result is measured. Also, in the same water system, two or more parameters that are any one of the water quality parameter, the control parameter and the result parameter are measured. A plurality of data sets of these pre-measurement results or pre-measurement indicators and parameters are prepared so that the pre-measurement results or pre-measurement indicators and parameters are varied by, for example, changing the day or time of measurement.
  • the pre-measured result or pre-measured index is assumed to be a function of two or more parameters, compared with the pre-measured result or pre-measured index to determine the form and coefficients of the function, and build a relationship model.
  • regression analysis methods linear model, generalized linear model, generalized linear mixed model, ridge regression , Lasso regression, elastic net, support vector regression, projection pursuit regression, etc.
  • time series analysis VAR model, SVAR model, ARIMAX model, SARIMAX model, state space model, etc.
  • decision tree decision tree, regression tree, random forest, XGBoost, etc.
  • neural networks simple perceptron, multilayer perceptron, DNN, CNN, RNN, LSTM, etc.
  • Bayes naive Bayes, etc.
  • clustering k-means, k-means++, etc
  • the relationship model is preferably a model obtained by regression analysis of the pre-confirmation result corresponding to the expected result or the index related to the pre-confirmation result and two or more parameters. Note that the number of sample sets for regression analysis is not particularly limited.
  • the relationship model in the same water system as the one for estimating the expected results. Also, for example, when the water quality of the water system changes greatly even within the same apparatus (for example, when the pulp that is the raw material for paper manufacturing is changed in the papermaking system of a paper mill), the water system after the water quality changes It is preferable to create and use a relationship model for .
  • the expected results or related indicators and two or more parameters are measured regularly or irregularly, and a relationship model is created each time, or data is added to the relationship model. may be updated.
  • the relationship model creation unit 25 is not an essential component, the relationship model may be created manually (manually) by, for example, an operator.
  • the second relationship model creating section 26 creates a second relationship model.
  • This second relationship model may be acquired by the second relationship model information acquisition unit 26 (to be described later) and may be used for estimating expected results by the second estimating unit 24 .
  • the second relationship model is a model that shows the relationship between the related index and the expected result.
  • the relationship model creation unit 25 will be called the "first relationship model creation unit" for convenience.
  • the second relationship model is not particularly limited, but includes, for example, a function indicating the relationship between the related index and the expected result, a lookup table, or a learned model of the relationship between the related index and the expected result.
  • the second relationship model is created, for example, as follows. Prior to inferring a prospective outcome, a prior measurement corresponding to the prospective outcome and a prior measurement indicator associated with the prior outcome are measured. A plurality of data sets of these preliminary measurement results and preliminary measurement indices are prepared so that the preliminary measurement results and the preliminary measurement indices are varied by, for example, changing the day and time of the measurement. An inferential model is then constructed using the pre-measured results as a function of the pre-measured index. Alternatively, for example, after preparing multiple data sets of pre-measurement results and pre-measurement indicators, thresholds are set for related indicators at points (pre-measurement indicators) where the pre-measurement results change significantly, and the second relationship model is generated. may be constructed.
  • the second relationship model in the same water system as the one for estimating the expected results. Also, for example, when the water quality of the water system changes greatly even within the same apparatus (for example, when the pulp that is the raw material for paper manufacturing is changed in the papermaking system of a paper mill), the water system after the water quality changes Preferably, a second relationship model for is created and used.
  • the second relationship model creating unit 26 is not an essential component, and the second relationship model may be created manually (manually) by, for example, an operator.
  • the second relationship model information acquisition unit 27 acquires the second relationship model.
  • the second relationship model may be one created by the second relationship model creating unit 26 .
  • the relationship model acquisition unit 22 is called the "first relationship model acquisition unit" for convenience.
  • the inference system 1 and the inference device 2 may include a relationship model evaluation unit (not shown).
  • the relationship model evaluation unit evaluates the relationship model created by the relationship model creation unit 25, and evaluates the degree of influence of each parameter information on the expected result or related index.
  • the relationship model evaluation unit evaluates the magnitude of the impact of each parameter information.
  • the method of evaluating the magnitude of the influence of each parameter information is not particularly limited. For example, when the expected result is expressed as a linear function of each parameter, the absolute value of the coefficient can be compared and evaluated.
  • the parameter information may be arranged in order of influence on the relationship model, and other than a predetermined number of parameter information may be excluded in descending order of influence, or a predetermined number of parameter information may be excluded in descending order of influence. may be provided, and parameters below this threshold may be excluded by the relationship model adjustment unit.
  • the inference system 1 and the inference device 2 may include a relationship model information adjustment unit (not shown).
  • the relationship model information adjustment unit performs adjustment to exclude parameter information that has a small impact on the relationship model, and then again instructs the relationship model information creation unit 25 to create relationship model information.
  • the evaluation of the relationship model evaluation unit and the adjustment of the relationship model information adjustment unit may be performed only once, or may be repeated twice or more. you can go
  • the output unit 3 is configured to output at least one of the expected result or related index calculated by the estimating unit 23 and the expected result estimated by the second estimating unit 24 .
  • the output unit 3 may, for example, display expected results or related indicators over time (expected results or related indicators vs. time graph, etc.).
  • the output unit 3 may output a warning when, for example, the expected result or related index exceeds a certain threshold.
  • the parameter information measurement unit 4 measures water quality parameters, control parameters, or result parameters.
  • parameter information measuring device 4 Although only one parameter information measuring device 4 is shown in FIG. 1 for convenience, it is not limited to this example, and two or more parameter information measuring devices may be used.
  • Various sensors can be selected as the measuring device, depending on the content of the parameters to be measured.
  • Examples of measuring devices include pH meter, electrical conductivity meter, oxidation-reduction potential meter, turbidity meter, thermometer, level meter for measuring foam height, COD meter, UV meter, particle size distribution meter, cohesion sensor, digital Camera (or digital video camera), internal bubble sensor, absorption photometer, freeness meter, dissolved oxygen meter, zeta potential meter, residual chlorine meter, hydrogen sulfide meter, retention/freeness meter, color sensor, hydrogen peroxide meter, etc. can be used.
  • Control parameters and the like may be directly input for controlling the device and may be used as they are. Such data may be communicated and received from the device. For this purpose, it may be recorded on a device other than the device.
  • the target water system of the estimation system is not particularly limited, and may be, for example, a water system in the process of manufacturing paper products. Specifically, if it is a process for manufacturing paper products, it includes a cooking process, a washing process, a black liquor concentration process, a causticizing process, and the like.
  • the target water system may be any water system other than the process of manufacturing paper products, such as various pipes, heat exchangers, storage tanks, kilns, washing equipment, and the like.
  • FIG. 3 is a schematic diagram showing the hardware configuration of the inference device according to this embodiment.
  • the guessing device 2 has a communication unit 51 , a storage unit 52 and a control unit 53 , and these components are electrically connected via a communication bus 54 inside the guessing device 2 . It is connected to the. These components are further described below.
  • the communication unit 51 is preferably a wired communication means such as USB, IEEE1394, Thunderbolt, wired LAN network communication, etc., but wireless LAN network communication, mobile communication such as 3G/LTE/5G, Bluetooth (registered trademark) communication, etc. is required. can be included depending on That is, it is more preferable to implement as a set of these communication means. As a result, information and instructions are exchanged between the guessing device 2 and other devices that can communicate with each other.
  • the storage unit 52 stores various information defined by the above description. This can be, for example, a storage device such as a solid state drive (SSD), or a random access memory (Random Access Memory: RAM) or the like. Moreover, the memory
  • the storage unit 52 also stores various programs that can be read by the control unit 53, which will be described later.
  • the control unit 53 processes and controls overall operations related to the inferring device 2 .
  • the control unit 53 is, for example, a central processing unit (CPU, not shown).
  • the control unit 53 implements various functions related to the guessing device 2 by reading a predetermined program stored in the storage unit 52 .
  • information processing by software stored in the storage unit 52
  • hardware control unit 53
  • FIG. 3 shows a single control unit 53, the present invention is not limited to this in practice, and a plurality of control units 53 may be provided for each function.
  • a single controller and multiple controllers may be combined.
  • the water quality parameter x, water quality parameter y, and control parameter z shown in Table 1 below are measured, and the number of trouble occurrences A is also measured to obtain a total of 30 sets of data. The data obtained are shown in Table 1 below.
  • the index a related to the expected result is represented by the following formula (1), where x, y, and z are the “parameters”, bn is the coefficient of x, y, and z, and a 0 and b 0 are constants.
  • FIG. 4 is a plot of the number of occurrences of trouble A for a total of 30 sets of data sets versus an index a related to the expected results.
  • FIG. 5 is a graph showing the magnitude of the influence of each parameter on the index a related to the expected result.
  • the control parameter z, the water quality parameter x, and the water quality parameter y have a greater influence on the index a related to the expected result in that order.
  • a negative binomial regression analysis using the standardized scores for each parameter yields the results in FIG.
  • the standardized score can be obtained by (individual numerical value ⁇ average value)/standard deviation.
  • the function of the index a related to the expected result used in the regression analysis, the water quality parameter x, the water quality parameter y, and the control parameter z is not limited to the above formula (1), and the general formula (2) can be used. can.
  • the estimation system 1 and the estimation device 2 as described above, it is possible to quantitatively estimate the occurrence of troubles in the water system W and the quality of products manufactured through the water system W. In particular, even if there are many parameters that affect the occurrence of troubles and product quality, it is possible to predict the occurrence of troubles and product quality by more accurately considering the respective influences.
  • the speculation program according to the present embodiment is a speculation program for estimating expected results that may occur in the water system or derived from the water system. Specifically, this estimation program causes a computer to function as a parameter information acquisition section, a relationship model information acquisition section, and an estimation section.
  • the parameter information acquisition unit acquires water quality parameters related to the water quality of the water system, control parameters related to the control conditions of equipment related to the water system or raw materials added to the water system, and parameters that have different meanings from the expected results and are related to the water system and the water system Parameter information that includes two or more parameters that are any one of the result parameters related to the results generated in the equipment or raw materials added to the water system, or derived from the water system, the equipment related to the water system, or the raw materials added to the water system to obtain.
  • the relationship model information acquisition unit acquires relationship model information indicating a relationship between an expected result or an index related to the expected result created in advance and two or more parameters.
  • the estimation unit estimates an expected result or an index related to the expected result based on the parameter information and the relationship model information.
  • parameter information acquisition unit the relationship model information acquisition unit, and the estimation unit can be similar to those of the estimation system described above, so descriptions thereof will be omitted here.
  • the estimating method is an estimating method for estimating expected results that may occur in the future in or derived from the water system.
  • this estimation method includes a parameter information acquisition process, a relationship model information acquisition process, and an estimation process.
  • the parameter information acquisition process water quality parameters related to the water quality of the water system, control parameters related to the control conditions of equipment related to the water system or raw materials added to the water system, and parameters that have different meanings from the expected results and are related to the water system
  • Parameter information that includes two or more parameters that are any one of the result parameters related to the results generated in the equipment or raw materials added to the water system, or derived from the water system, the equipment related to the water system, or the raw materials added to the water system get.
  • relationship model information acquisition step relationship model information indicating the relationship between the expected result created in advance or an index related to the expected result and two or more parameters is acquired.
  • estimating step an expected result or an index associated with the expected result is estimated based on the parameter information and the relationship model information.
  • FIG. 6 is a flowchart of the estimation method according to this embodiment.
  • parameter information is acquired (parameter information acquisition step S1)
  • relationship model information is acquired (relationship model information acquisition step S2)
  • these are input.
  • an expected result or an index related to the expected result is guessed (estimating step S3).
  • Example 1 In the water system of the paper production facility (a continuous water system consisting of the raw material system, the papermaking system, and the recovery system), the redox potential of the raw material system 1, the redox potential of the raw material system 2, and the redox potential, turbidity, and pH of the papermaking system. , water temperature, and foam height in the recovery system were measured as water quality parameters, respectively, and the average values for 24 hours before the production of the corresponding paper products were used.
  • 7 is a schematic diagram of equipment for manufacturing paper in Example 1.
  • defect index (sometimes called “ ⁇ ”)
  • ⁇ the number of defects occurring within 24 hours
  • seven water quality parameters and one result parameter function (hereinafter referred to as "defect index (sometimes called “ ⁇ ”)) relationship model was created. More specifically, as a procedure for creating the defect index ⁇ , after excluding parameters that are two standard deviations or more away from the average value as outliers, regression analysis was performed using IBM's SPSS Modeler. The correlation coefficient between the defect index ⁇ obtained by regression analysis and the number of defects was 0.71 (p ⁇ 0.05), indicating a strong correlation.
  • the water quality parameters and parameters are applied to the above-mentioned relationship model was applied to calculate the defect index ⁇ , and the correlation coefficient with the number of defects was calculated to be 0.71 (p ⁇ 0.05). From this, it was confirmed that there is a strong correlation between the number of defects and the defect index ⁇ , and that the defect index ⁇ is effective in predicting the number of defects that will occur in the future.
  • FIG. 8 is a plot of the number of defects versus defect index ⁇ for a total of 572 data sets in Example 1.
  • FIG. 8 372 sets of data sets for relationship model creation are indicated by circles, and 200 sets of data sets for accuracy verification are indicated by rectangles.
  • FIG. 9 is a graph showing the degree of influence of each parameter on the defect index ⁇ .
  • Raw material system 1 to 3 of the corrugated board (liner) production facility pH, electrical conductivity, raw material system 2 pH, oxidation-reduction potential, raw material system 3 pH, electrical conductivity, papermaking system: water temperature, electrical conductivity were measured as water quality parameters, respectively (see FIG. 7).
  • water quality parameter measurement we also measured the basic unit of paper strength agent usage of manufactured paper products, and prepared 60 sets of these data sets. These data sets were randomly divided into 7:3, and 70% (42 pairs) were used as relationship model creation data and 30% (18 pairs) as model validation data.
  • a relationship model that indicates the reciprocal of the paper strength agent usage unit was created as a "paper strength index" as a function of the eight water quality parameters described above. More specifically, as a procedure for creating a paper strength index, after excluding parameters that are two standard deviations or more away from the average value as outliers, regression analysis was performed using IBM's SPSS Modeler. The correlation coefficient between the paper strength index obtained by regression analysis and the basic unit of paper strength agent usage was ⁇ 0.58 (p ⁇ 0.05), indicating a correlation.
  • FIG. 10 is a plot of paper strength agent usage intensity vs. paper strength index value for a total of 60 sets of data sets in Example 2.
  • 42 sets of data sets for relationship model creation are indicated by circles, and 18 sets of data sets for accuracy verification are indicated by rectangles.
  • FIG. 11 is a graph showing the magnitude of the influence of each parameter on the paper strength index.
  • the paper strength index proportional to the amount of paper strength used per unit
  • Example 3 Temperature, pH, oxidation-reduction potential, electrical conductivity, turbidity, standing turbidity, pH, Oxidation-reduction potential, electrical conductivity, turbidity, and recovery system turbidity were measured as water quality parameters (see FIG. 7). Operation timing, papermaking speed, amount of internal additives added, felt moisture content, ash content in paper, product basis weight, and product brand were used as control parameters. Furthermore, the paper break timing was measured, and 138,276 sets of these data sets were prepared. The same water quality parameters, control parameters, and paper break timing were used.
  • paper breakage index a relationship model for Specifically, as a procedure for creating a paper break occurrence index, parameters that are two standard deviations or more away from the average value are excluded as outliers, and regression analysis is performed using IBM's SPSS Modeler to determine the relationship created a gender model. Since the number of data sets is enormous, plot diagrams are omitted.
  • FIGS. 12 and 13 are graphs showing the occurrence of paper breakage and the temporal change of the paper breakage index.
  • the vertical axis indicates occurrence of paper breakage and the paper breakage index
  • the horizontal axis indicates time. Circles indicate the presence or absence of actual paper breakage (0 indicates no paper breakage, 1 indicates paper breakage occurrence) (left axis of ordinate), and squares indicate the index of paper breakage (right axis of ordinate). ). From FIGS. 12 and 13, it is clear that a paper break actually occurs when the paper break index approaches one. It was found that a paper break can be predicted by appropriately setting a threshold value for the paper break index.
  • Example 4 In the water system of the paper production facility (continuous water system consisting of raw material system, papermaking system, recovery system, and drainage system), the pH and turbidity of the raw material system 1, the redox potential of the papermaking system, and the electrical conductivity of the drainage system are measured. It was measured as a water quality parameter, and the value measured 16 hours before the production of the corresponding paper product was used. 14 is a schematic diagram of equipment for manufacturing paper in Example 4. FIG. In addition, the speed of paper products was measured as a control parameter. In addition, the basis weight of the paper product was measured as a result parameter. In addition, the number of defects in paper products was measured, and 647 sets of these data sets were prepared.
  • defect index ⁇ a relationship model of the number of defects and functions of four water quality parameters, one control parameter and one result parameter (hereinafter sometimes referred to as "defect index ⁇ ") was created. More specifically, as a procedure for creating the defect index ⁇ , after excluding parameters that are two standard deviations or more away from the average value as outliers, use the R language package KFAS, which is a programming language for statistical analysis. Then, we performed an analysis using a state-space model. The correlation coefficient between the defect index ⁇ obtained by the state space model and the number of defects was 0.62 (p ⁇ 0.05), confirming that there is a correlation.
  • FIG. 15 is a plot of the number of defects versus the defect index ⁇ value of the data set for creating the data of the total of 647 sets in Example 4.
  • FIG. 16 is a plot of the number of defects in the total 255 data sets for accuracy verification in Example 4 versus the defect index ⁇ value for data creation.
  • FIG. 17 is a plot showing changes over time in defect indices calculated from a total of 255 sets of data sets for accuracy verification in Example 4.
  • FIG. 18 is a graph showing the degree of influence of each parameter on the defect index ⁇ .
  • Example 5 The pH and turbidity of the water system of the paper production facility (a continuous water system consisting of the raw material system, the papermaking system, the recovery system, and the drainage system), the redox potential of the papermaking system, and the electrical conductivity of the wastewater system are measured. It was measured and used as a parameter (see FIG. 14). In addition, the speed of paper products was measured as a control parameter. In addition, the basis weight of the paper product was measured as a result parameter. In addition, the number of defects in paper products was measured, and 631 sets of these data sets were prepared.
  • defect index ⁇ a relationship model of the number of defects and functions of four water quality parameters, one control parameter and one result parameter (hereinafter sometimes referred to as "defect index ⁇ ") was created. More specifically, as a procedure for creating the defect index ⁇ , after excluding parameters that are two standard deviations or more away from the average value as outliers, the R language package vars, which is a programming language for statistical analysis, is used. We conducted an analysis using the VAR model, which is a type of time series analysis. It was confirmed that the defect index ⁇ obtained by the VAR model and the number of actually generated defects are linked.
  • FIG. 19 is a plot of the number of defects versus the defect index ⁇ value of a total of 631 sets of data sets for relationship model creation.
  • FIG. 20 is a plot of defect count versus defect index ⁇ value for a total of 255 datasets for accuracy verification in Example 5;
  • FIG. 21 is a plot showing changes over time in defect indices calculated from a total of 271 sets of data sets for accuracy verification in Example 5.
  • FIG. 21 is a plot showing changes over time in defect indices calculated from a total of 271 sets of data sets for accuracy verification in Example 5.
  • FIG. 22 is a graph showing the degree of influence of each parameter on the defect index ⁇ . Note that FIG. 22 shows only parameters that are significant (p ⁇ 0.10) at a significance level of 10%. When considering a process to reduce the number of defects, it is necessary to specify parameters that have a large effect on the defect index ⁇ (proportional to the number of defects), and to prioritize investigation and improvement of the causes of fluctuations. It is also possible to effectively reduce the number of defects.
  • Example 6 Measure the pH and turbidity of the raw material system, the oxidation-reduction potential of the papermaking system, and the electrical conductivity of the drainage system as water quality parameters in the water system of the paper production facility (a continuous water system consisting of the raw material system, the papermaking system, and the drainage system). 16 hours before production of the corresponding paper product was used (see FIG. 14). In addition, the speed of paper products was measured as a control parameter. In addition, as result parameters, the basis weight of paper products and the number of defects of paper products were measured, and 1706 sets of these data sets were prepared.
  • the number of defects that occur after 16 hours, four water quality parameters, one control parameter and one result parameter function (below , sometimes called “defect index ⁇ ”). More specifically, as a procedure for creating the defect index ⁇ , after excluding parameters that are two standard deviations or more away from the average value as outliers, IBM's SPSS Modeler is used to create a multi-layer perceptron, which is a type of neural network. Analysis was performed by The correlation coefficient between the defect index ⁇ obtained by the multi-layer perceptron and the number of defects was 0.73 (p ⁇ 0.05), indicating a strong correlation.
  • the water quality parameters and parameters are applied to the above-mentioned relationship model was applied to calculate the defect index ⁇ , and the correlation coefficient with the number of defects was calculated to be 0.73 (p ⁇ 0.05). From this, it was confirmed that there is a strong correlation between the number of defects and the defect index ⁇ , and that it is effective for predicting the number of defects that will occur in the future.
  • FIG. 23 is a plot of the number of defects versus the defect index ⁇ value of a total of 1216 sets of data sets for relationship model creation in Example 6.
  • FIG. 24 is a plot of the number of defects versus the defect index ⁇ value for a total of 490 sets of data sets for accuracy verification in Example 6;
  • FIG. 11 is a plot of the number of defects versus the defect index ⁇ value for the data set of Example 6;
  • FIG. 23 is a plot of the number of defects versus the defect index ⁇ value of a total of 1216 sets of data sets for relationship model creation in Example 6.
  • FIG. 24 is a plot of the number of defects versus the defect index ⁇ value for a total of 490 sets of data sets for accuracy verification in Example 6;
  • FIG. 11 is a plot of the number of defects versus the defect index ⁇ value for the data set of Example 6;
  • FIG. 25 is a graph showing the degree of influence of each parameter on the defect index ⁇ .
  • Example 7 Measure the pH and turbidity of the raw material system, the oxidation-reduction potential of the papermaking system, and the electrical conductivity of the drainage system as water quality parameters in the water system of the paper production facility (a continuous water system consisting of the raw material system, the papermaking system, and the drainage system). 16 hours before production of the corresponding paper product was used (see FIG. 14). In addition, the flow rate of the seed box and the speed of the paper product were measured as control parameters. In addition, as result parameters, the basis weight of the paper product and the number of defects of the paper product were measured, and 2040 sets of these data sets were prepared.
  • the number of defects occurring after 16 hours four water quality parameters, two control parameters and one result parameter function (below , sometimes called “defect index ⁇ ”). More specifically, as a procedure for creating the defect index ⁇ , after excluding parameters that are two standard deviations or more away from the average value as outliers, IBM's SPSS Modeler is used to determine a decision tree and a kind of ensemble learning. An XGBoost analysis was performed. The correlation coefficient between the defect index ⁇ obtained by XGBoost and the number of defects was 0.95 (p ⁇ 0.05), indicating a strong correlation.
  • the water quality parameters and parameters are applied to the above-mentioned relationship model was applied to calculate the defect index ⁇ , and the correlation coefficient with the number of defects was calculated to be 0.57 (p ⁇ 0.05). From this, it was confirmed that there is a correlation between the number of defects and the defect index ⁇ , and that it is effective for predicting the number of defects that will occur in the future.
  • FIG. 26 is a plot of the number of defects versus the defect index ⁇ of a total of 1503 sets of data sets for relationship model creation in Example 7.
  • FIG. FIG. 27 is a plot of defect number versus defect index ⁇ for a total of 537 datasets for accuracy verification in Example 7;
  • FIG. 28 is a graph showing the degree of influence of each parameter on the defect index ⁇ .

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Pathology (AREA)
  • Medicinal Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Food Science & Technology (AREA)
  • Emergency Medicine (AREA)
  • Biomedical Technology (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Molecular Biology (AREA)
  • Mechanical Engineering (AREA)
  • Manufacturing & Machinery (AREA)
  • Hydrology & Water Resources (AREA)
  • Environmental & Geological Engineering (AREA)
  • Water Supply & Treatment (AREA)
  • Organic Chemistry (AREA)
  • Paper (AREA)
  • Testing And Monitoring For Control Systems (AREA)
PCT/JP2022/004611 2021-02-12 2022-02-07 推測装置、推測システム、推測プログラム及び推測方法 WO2022172882A1 (ja)

Priority Applications (3)

Application Number Priority Date Filing Date Title
KR1020237027836A KR20230133886A (ko) 2021-02-12 2022-02-07 추측장치, 추측 시스템, 추측 프로그램 및 추측방법
EP22752706.6A EP4286583A1 (en) 2021-02-12 2022-02-07 Estimation device, estimation system, estimation program, and estimation method
CN202280014302.7A CN116829785A (zh) 2021-02-12 2022-02-07 推测装置、推测系统、推测程序以及推测方法

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2021021053 2021-02-12
JP2021-021053 2021-02-12

Publications (1)

Publication Number Publication Date
WO2022172882A1 true WO2022172882A1 (ja) 2022-08-18

Family

ID=82837547

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/004611 WO2022172882A1 (ja) 2021-02-12 2022-02-07 推測装置、推測システム、推測プログラム及び推測方法

Country Status (6)

Country Link
EP (1) EP4286583A1 (zh)
JP (2) JP2022123880A (zh)
KR (1) KR20230133886A (zh)
CN (1) CN116829785A (zh)
TW (1) TW202302960A (zh)
WO (1) WO2022172882A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117303518A (zh) * 2023-09-08 2023-12-29 深圳市伊科赛尔环保科技有限公司 一种电离子交换超纯水设备及其控制方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03238346A (ja) * 1990-02-16 1991-10-24 Nippon Kamiparupu Kenkyusho:Kk パルプまたは紙の製造工程におけるピッチの定量法
US20030045962A1 (en) * 2001-08-30 2003-03-06 Evren Eryurek Control system using process model
WO2003074784A1 (en) * 2002-03-07 2003-09-12 University Of Manchester Institute Of Science & Technology A method of dynamically modelling a paper manufacturing plant using pca (principal component analysis)
WO2012070644A1 (ja) 2010-11-25 2012-05-31 栗田工業株式会社 紙を製造する方法
JP2019193916A (ja) * 2018-05-01 2019-11-07 株式会社東芝 凝集剤注入制御装置、凝集剤注入制御方法及びコンピュータプログラム

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120070644A (ko) 2010-12-22 2012-07-02 두산인프라코어 주식회사 건설중장비의 원격관리를 위한 다중 스케줄링 방법

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03238346A (ja) * 1990-02-16 1991-10-24 Nippon Kamiparupu Kenkyusho:Kk パルプまたは紙の製造工程におけるピッチの定量法
US20030045962A1 (en) * 2001-08-30 2003-03-06 Evren Eryurek Control system using process model
WO2003074784A1 (en) * 2002-03-07 2003-09-12 University Of Manchester Institute Of Science & Technology A method of dynamically modelling a paper manufacturing plant using pca (principal component analysis)
WO2012070644A1 (ja) 2010-11-25 2012-05-31 栗田工業株式会社 紙を製造する方法
JP2019193916A (ja) * 2018-05-01 2019-11-07 株式会社東芝 凝集剤注入制御装置、凝集剤注入制御方法及びコンピュータプログラム

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117303518A (zh) * 2023-09-08 2023-12-29 深圳市伊科赛尔环保科技有限公司 一种电离子交换超纯水设备及其控制方法
CN117303518B (zh) * 2023-09-08 2024-06-04 深圳市伊科赛尔环保科技有限公司 一种电离子交换超纯水设备及其控制方法

Also Published As

Publication number Publication date
EP4286583A1 (en) 2023-12-06
JP2022123880A (ja) 2022-08-24
KR20230133886A (ko) 2023-09-19
TW202302960A (zh) 2023-01-16
CN116829785A (zh) 2023-09-29
JP2024037736A (ja) 2024-03-19

Similar Documents

Publication Publication Date Title
JP2024037736A (ja) 推測装置、推測システム、推測プログラム及び推測方法
KR20130102608A (ko) 종이를 제조하는 방법
AU2018379543B2 (en) Method for predicting or controlling microbial status of a paper or board making process
US7181308B2 (en) Method for monitoring and analyzing a paper production process
JP2022121356A (ja) 汚泥処理設備運転支援ナビゲーションシステム、汚泥処理設備運転支援方法
WO2023127748A1 (ja) 異常検知装置、異常検知方法及び異常検知プログラム
WO2022163549A1 (ja) 推定装置、予測装置、制御装置、推定システム、予測システム、制御システム、推定プログラム、予測プログラム、制御プログラム、推定方法、予測方法及び制御方法
WO2022168965A1 (ja) 汚泥処理設備運転支援ナビゲーションシステム、汚泥処理設備運転支援方法
JP2022123880A5 (zh)
CN115508322A (zh) 一种水污染源多维度溯源方法及装置
JP5541790B2 (ja) プラント運転監視装置およびシステムならびにプログラム
WO2024034509A1 (ja) 廃液処理設備の運転方法、処理後廃液の性質を予測する方法、運転システム及び予測システム
TW202414321A (zh) 廢液處理設備的運轉方法、預測處理後廢液的性質的方法、運轉系統以及預測系統
JP2024086874A (ja) 廃液処理設備の運転方法、処理後廃液の性質を予測する方法、運転システム及び予測システム
JPS6047958B2 (ja) 懸濁液の連続洗浄方法
US20220251781A1 (en) Process to optimize brown stock washing unit operations
Moguel-Castañeda et al. Diagnosis of the Cane Sugar Crystallization Process by Multifractal Analysis of Temperature Time Series
WO2020189301A1 (ja) 運転方針評価装置、運転方針評価方法、およびプログラム
Heikkinen et al. SOM-based modelling for an activated sludge treatment process
US20240192676A1 (en) A method for estimating disturbances and giving recommendations for improving process performance
US20220326685A1 (en) Estimating risk level in an aqueous process
JP2023180657A (ja) 白液処理方法、白液酸化監視システム
Zhang Fuzzy Mean Clustering Analysis Based on Glutamic Acid Fermentation Failure
CN117521008A (zh) 一种基于数据融合的污水处理监测控制方法及系统
MIC 1. Predictive Microbiology

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22752706

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 202280014302.7

Country of ref document: CN

ENP Entry into the national phase

Ref document number: 20237027836

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 1020237027836

Country of ref document: KR

ENP Entry into the national phase

Ref document number: 2022752706

Country of ref document: EP

Effective date: 20230831

NENP Non-entry into the national phase

Ref country code: DE